Combined effects of oligofructose and Bifidobacterium animalis on gut microbiota and glycemia in obese rats

Authors


  • Fundig agencies: This project was funded by Canadian Institutes of Health Research (MOP 115076). MRB is supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Alberta Innovates Health Solutions (AIHS). DCS is supported by AIHS and a CIHR Training Program Fellowship from Alberta Children's Hospital Research Institute for Child and Maternal Health. DTR is supported by an NSERC scholarship. HAP is supported by AIHS and NSERC.

  • Disclosure: MR Bomhof, DC Saha, DT Reid, and HA Paul declare no conflict interest. RA Reimer previously held a research grant from Beneo-Orafti Inc., manufacturer of Raftilose P95, for a project unrelated to the current work.

  • Author Contributions: MRB carried out experiments, collected data, analyzed data, generated figures, and wrote paper. DCS, DTR, and HAP carried out experiments. RAR designed research, analyzed data, wrote paper, and had primary responsibility for final content. All authors read and approved the final manuscript.

Abstract

Objective

Prebiotics and probiotics may be able to modify an obesity-associated gut microbiota. The aim of this study was to examine the individual and combined effects of the prebiotic oligofructose (OFS) and the probiotic Bifidobacterium animalis subsp. lactis BB-12 (BB-12) on gut microbiota and host metabolism in obese rats.

Methods

Adult male, diet-induced obese Sprague Dawley rats were randomized to: (1) Control (C); (2) 10% OFS; (3) BB-12; (4) OFS + BB-12 for 8 weeks (n = 9-10 rats/group). Body composition, glycemia, gut permeability, satiety hormones, cytokines, and gut microbiota were examined.

Results

Prebiotic, but not probiotic reduced energy intake, weight gain, and fat mass (P < 0.01). OFS, BB-12, and the combined OFS + BB-12 improved glycemia (P < 0.05). Individually, OFS and BB-12 reduced insulin levels (P < 0.05). Portal GLP-1 was increased with OFS, whereas probiotic increased GLP-2 (P < 0.05). There was a marked increase in bifidobacteria and lactobacilli (P < 0.01) with OFS that was not observed with probiotic alone.

Conclusions

The impact of prebiotic intake on body composition and gut microbiota was of greater magnitude than the probiotic BB-12. Despite this, an improvement in glucose AUC with both prebiotic or probiotic demonstrates the beneficial role of each of these “biotic” agents in glycemic control.

Introduction

The human biological system, while capable of coping with food scarcity, proves vulnerable to chronic food excess [1]. Among the many factors that influence the development of obesity, there is a growing body of research demonstrating the prominent role of gut microbiota in energy balance and metabolism. Through centuries of co-evolution, a complex ectosymbiotic relationship has developed between gut microbiota and humans [2]. In recent years it has been shown that gut microbiota differ between the lean and obese states [3]. It is theorized that in the context of obesity, the symbiotic relationship between gut microbiota and host shifts to a “dysbiotic” state, such that the microbiota impair host metabolism. Some but not all reports have shown that obesity is characterized by an increased ratio of Firmicutes to Bacteroidetes and reduced bifidobacteria cell numbers [3-5].

While fulfilling important biological and metabolic functions for the host, the microbiome offers a component of the human metagenome that can be altered through minimally invasive means. Strategies to restore gut microbiota include the use of prebiotics and probiotics. Prebiotics are nondigestible food ingredients that beneficially alter the composition and metabolism of the gut microbiota in such a way that it confers a health benefit to the host [6]. Oligofructose (OFS) is highly fermentable in the caecal-colon and has been demonstrated to selectively enhance Bifidobacterium spp. and Lactobacillus spp. in favor of a lean phenotype [7, 8]. Probiotics are defined as “live micro-organisms which when administered in adequate amounts confer a health benefit on the host” [9]. Mounting evidence suggests that probiotics, such as Bifidobacterium spp. and Lactobacillus spp. are able to reduce adiposity, cholesterol, and body weight as well as improve glycemia [10-15].

The combination of a prebiotic and a probiotic is referred to as a synbiotic. Several studies have suggested that synbiotics have antiobesogenic properties [16, 17]. These studies, however, did not compare the effect of a synbiotic to a prebiotic or probiotic alone and conclusions about synergistic potential could not be evaluated. Our objective was to determine the synergistic, as well as singular effects and mechanisms by which the prebiotic OFS and the probiotic Bifidobacterium animalis subsp. lactis BB-12 alter gut microbiota profiles, glycemia, satiety hormone secretion, and adiposity in obese rats.

Methods

Animals and treatments

The study protocol was approved by The University of Calgary Animal Care Committee and conformed to the Guide for the Care and Use of Laboratory Animals. Eighty male Sprague-Dawley rats (10 wks of age) were obtained from Charles River (Charles River, St. Constant, PQ) and housed three per cage on a 12-hr light–dark cycle in a temperature (20-22°C) and humidity controlled (41-60%) room. Rats were fed a high-fat/sucrose diet ad libitum for eight weeks (Dyet, Bethlehem, PA) to induce obesity. The 40 rats with the greatest weight gain were transitioned to individual housing and randomized into 1 of 4 groups: [1] Control (C, AIN-93M diet); [2] 10% (wt/wt) OFS (Orafti P95, BENEO-Orafti Inc.); [3] B. animalis subsp. lactis BB-12® (Chr. Hansen, Milwaukee, WI) (1 × 1010 CFU/d) (BB-12); [4] 10% OFS + BB-12 (OFS + BB-12) for a period of eight weeks (n = 10 rats/group). The 10% dose of OFS was selected based on previous literature showing reductions in fat mass at this level [18]. The dose of BB-12 was selected based on previous studies in rodents using a similar dose and evidence that it survives gastrointestinal transit in rats [19]. Composition of experimental diets is provided in Supporting Information Table S1. Diets containing the probiotic B. lactis BB-12 were prepared fresh three times weekly. Freeze-dried probiotic BB-12 powder (3 × 1010 CFU/g) was mixed with AIN-93M diet to ensure that each rat received 1 × 1010 CFU/d BB-12 daily (∼1.2-1.7 g/100 g AIN-93M). Body weight was measured once every week and food intake was measured three times per week. One day prior to sacrifice, rats were lightly anaesthetized with isoflurane and body composition was measured via DXA scan with software for small animals (Hologic ODR 4500; Hologic).

In vivo intestinal permeability

Based on methods previously described [20, 21], movement of fluorescein isothiocyanate-dextran-4000 daltons (FD-4) (Sigma-Aldrich, St. Louis, MO, USA) across the epithelium was used to assess intestinal permeability. On week 6, following 16 h food deprivation, rats were gavaged with 500 mg/kg FD-4. At 1 and 4 h postgavage, 300 µl of blood was collected from the tip of the tail and centrifuged at 4°C for 3 min (12,000g). Plasma samples were diluted in equal volumes of PBS and 100 μl loaded in duplicate into a 96 well plate. Standards were made by serial diluting FD-4 in a 1:3 v/v mixture of plasma:PBS. FD-4 concentrations were measured with a fluorescence spectrophotometer (excitation of 485 nm and emission of 535 nm).

Oral glucose tolerance test

One week prior to sacrifice, following 16 h feed deprivation, rats were given an oral gavage of 2 g/kg glucose. Blood was collected via tail nick at 0, 15, 30, 60, 90, and 120 min postgavage into a chilled tube containing diprotinin-A (0.034 mg/ml blood; (MP Biomedicals, Irvine, CA), Sigma protease inhibitor (1 mg/ml blood; Sigma Aldrich, Oakville, ON, Canada) and Roche Pefabloc (1 mg/ml of blood; Roche, Mississauga, ON, Canada). Blood glucose was measured immediately with a blood glucose meter (OneTouch Glucose Meter, Lifescan). Plasma was stored at −80°C until analysis for satiety hormones and cytokines. The insulinogenic index and composite insulin sensitivity index (CISI) was calculated as previously described [22]. The early phase and total insulin response for the insulinogenic index was calculated using the formula (Insulin AUC15)/(Glucose AUC15) and (Insulin AUC120)/(Glucose AUC120), respectively. CISI was calculated with the formula:

display math

Plasma and tissue samples

Following 16 h feed deprivation, rats were anaesthetized with isoflurane. A portal blood sample was collected with inhibitors described above. Following the blood draw, rats were killed by over-anesthetization and aortic cut. The liver, stomach, small intestine, cecum, and colon were excised, weighed, and snap frozen in liquid nitrogen. Fecal and cecal matters were collected. All samples were stored at −80°C until analysis.

Plasma satiety hormones, cytokines, and LPS

Portal plasma glucagon-like peptide-1(active), glucagon-like-peptide 2 (GLP-2), and adiponectin were measured using ELISA kits (Millipore, Billerica, MA). Ghrelin (active), amylin (active), insulin, leptin, glucose-dependant insulinotropic polypeptide (GIP) (total), GLP-1 (active), and PYY (total) were quantified using a Milliplex Rat Gut Hormone Panel. TNFα, MCP-1, IL-1β, PAI-1(total), and IL-6 concentrations were quantified using a Milliplex Rat Adipokine Panel (Millipore). Portal plasma LPS was measured using a PyroGene Recombinant Factor C Endotoxin Endpoint Fluorescent Detection assay (Lonza) according to manufacturer directions.

Hepatic triglyceride content

Triglycerides (TG) were measured using TG (GPO) reagent set (Point Scientific, Lincoln Park, MI, USA) according to our previous work [23].

Gut microbiota profiling using qPCR

Microbial profiling was performed according to our previous work [7]. Briefly, total bacterial DNA was extracted from fecal/cecal samples using FastDNA Spin Kit for Feces (MP Biomedicals, Lachine, QC, Canada) and quantified using PicoGreen DNA quantification kit (Invitrogen, Carlsbad, CA, USA). All samples were brought to a concentration of 4 ng/µl prior to storage at −20°C for later analysis. Amplification and detection were conducted in 96-well plates with SYBR Green 2 × qPCR Master Mix (BioRad). Samples were run in duplicate with a final volume of 25 µl containing 0.3 µM primer and 20 ng template gDNA. Group specific primers are provided in Supporting Information Table S2. The primers for Lactobacillus spp. do not cover the entire genera, but are specific for 15 species of lactobacilli (L. acidophilus, L. amylolyticus, L. amylovorus, L. crispatus, L. fornicalis, L. gallinarum, L. hamsteri, L. helveticus, L. intestinalis, L. jensenii, L. kefiranofaciens ssp.kefirgranum, L. kitasatonis, L. psittaci, L. suntoryeus, and L. ultunensis) [24]. The specificity of the primers and the limit of detection were determined according to Louie et al. [25]. The standard for B. animalis was obtained by extracting DNA from BB-12 freeze dried powder [26] and quantified. The 16S rRNA gene copies value was calculated according the following webpage: http://cels.uri.edu/gsc/cndna.html using average genome sizes. Standard curves were normalized to the copy number of the 16S rRNA gene obtained from the following webpage: http://rrndb.mmg.msu.edu/index.php.

Expression of genes related to intestinal permeability

Gene expression for tight junction protein-1 (TJP1) and occludin were analyzed according to previous real-time PCR work [27]. The primers were designed using Beacon Designer 3 software. TJP1: Sense Primer CCATGCCTCCTCCTCCTC, Anti-sense Primer ACG GAATTGCCTTCACTCTG; occludin: Sense Primer GAGGACTGG CTCAGGGAATATC, Anti-sense Primer TTGTTGACCTCGTCGA GTTCTG. Actin was confirmed to be a suitable reference gene and is not changed in response to the treatment. PCR amplification efficiency was established for actin, occludin, and TJP1 by means of calibration curves. The 2-ΔΔCt calculation was used to determine the relative difference in mRNA expression.

Determination of BB-12 viability

The concentration of B. lactis BB-12 supplied by Chr. Hansen was verified via traditional culture technique. Freeze dried BB-12 powder was suspended in autoclaved PBS containing 0.05% filter sterilized cysteine-HCl, followed by preparation of a 10-fold dilution series which was spread plated under anaerobic conditions. About 0.1 ml of each of the dilution series samples were plated on MRS agar plate supplemented with 0.05% cysteine-HCl. After 72 hr incubation, the plates were removed and the CFU were counted to verify the stated concentration of the product.

BB-12 gavage study

To determine whether the storage time and food matrices had an impact on the inoculation potential of the BB-12 a brief gavage study was conducted. Sixteen male Sprague-Dawley rats (10 wk of age) were randomized to receive a daily gavage of either: [1] PBS control (Gavage - C) or [2] 1 × 1010 CFU/d BB-12 suspended in PBS (Gavage BB-12) for two weeks while consuming AIN-93M diet ad libitum. Gut microbiota was analyzed as above.

Statistical analysis

All data are presented as mean ± SEM. A 2-way ANOVA was used to determine the main effects of prebiotic (OFS vs control-containing treatments), probiotic (BB-12 vs control-containing treatments) and their interaction. If a significant interaction was identified, a one way ANOVA with all four treatments was performed with Tukey's post hoc test to determine differences between experimental groups. For measurements with repeated measures, a 2-way repeated measures ANOVA was used with time as the within-subject factor and prebiotic and probiotic as between-subject factors. For comparisons between the treatment groups in the BB-12 gavage study a one-way ANOVA was used. Analysis was completed using SPSS V19.0 software (SPSS, Chicago, IL, USA). Data were considered significant at P < 0.05.

Results

Food intake, body composition, hepatic triglyceride, and cecum weight

Prebiotic but not probiotic reduced energy intake (P = 0.003; Table 1). While body weight increased with time in all rats (P < 0.001), only prebiotic reduced weight gain (P = 0.002) (Figure 1), fat mass (P = 0.003), and body fat % (P = 0.006) (Table 1). Bone mineral density (P = 0.013) and cecum weight (P < 0.001) was higher in rats consuming prebiotic. Prebiotic reduced hepatic TG (P = 0.001).

Figure 1.

Body weight of obese rats treated with OFS, BB-12, both or neither for 8 wk. Values are mean ± SEM, n = 9-10. BB-12, B. animalis ssp. lactis BB-12; C, control; OFS, oligofructose.

Table 1. Energy intake and body composition in obese rats treated with OFS, BB-12, both or neither for 8 wka
     Two way ANOVA P-values
TreatmentCOFSBB-12OFS + BB-12PreProPre × Pro
  1. a

    Values are means ± SEM, n = 9-10.

  2. b

    Refers to mean daily food intake over the 8 wk study. BB-12, B. animalis subsp. lactis BB-12; C, control; OFS, oligofructose; Pre, prebiotic; Pro, probiotic

Food intakeb, g/d23.9 ± 0.324.3 ± 0.423.5 ± 0.523.0 ± 0.50.920.070.34
Energy intake, kJ/d360.7 ± 5.0349.4 ± 5.7354.7 ± 7.8331.2 ± 6.50.010.070.37
Final body weight, g485 ± 7.6479 ± 6.9490 ± 6.8470 ± 7.20.080.760.36
Total weight change, g/8 wk60.0 ± 4.0552.6 ± 4.3465.1 ± 4.3956.3 ± 4.500.0030.770.23
Fat mass, g60.0 ± 5.2849.6 ± 3.5165.4 ± 4.6149.5 ± 2.950.0030.530.50
Percent fat, %12.3 ± 1.0510.4 ± 0.7613.3 ± 0.8810.5 ± 0.620.0060.490.61
BMD, g/cm30.169 ± 0.0010.178 ± 0.0020.174 ±0.0030.177 ± 0.0010.0130.420.25
Liver, g12.6 ± 0.4912.0 ± 0.3311.9 ± 0.2912.0 ± 0.730.670.480.51
Liver TG, μg/mg29.7 ± 0.9928.03 ± 0.6830.0 ± 1.2824.6 ± 0.580.0010.100.05
Cecum weight, g0.93 ± 0.051.91 ± 0.130.96 ± 0.072.11 ± 0.190.0010.350.48

Glycemic and insulinemic response

Glucose and insulin during the oral glucose tolerance test (OGTT) were affected by time (P < 0.001) and the interaction of time and prebiotic (P < 0.05; Figure 2A). The interaction between prebiotic and probiotic affected blood glucose levels during the OGTT (P = 0.024). At fasting, both OFS (P = 0.028) and OFS+BB-12 (P = 0.022) were lower than C. At 15 min, OFS + BB-12 (P = 0.009) was lower than C. At 90 min, both OFS (P = 0.007) and BB-12 (P = 0.003) were lower than C. Glucose AUC was affected by the interaction between prebiotic and probiotic (P = 0.012), whereby OFS, BB-12, and OFS+BB-12 (P < 0.015) were all lower than C. Independently, prebiotic (P = 0.019) and probiotic (P = 0.035) reduced insulin levels during the OGTT (Figure 2B). Only probiotic (P = 0.007) reduced fasting insulin concentrations. Insulin AUC was significantly reduced by prebiotic (P = 0.035) and probiotic (P = 0.048)(Table 2).

Figure 2.

Blood glucose (A), plasma insulin (B), amylin (C), leptin (D), ghrelin (E), GIP (F), PYY (G) and portal GLP-1 (H) of obese rats treated with OFS, BB-12, both or neither for 8 wk. Values are mean ± SEM, n = 9-10. Labeled means at a time without a common letter differ, P < 0.05. BB-12, B. animalis subsp. lactis BB-12; C, control; GIP, glucose-dependent insulinotropic polypeptide; GLP-1, glucagon-like peptide 1;OFS, oligofructose; PYY, peptide tyrosine tyrosine.

Table 2. AUC for blood glucose and plasma satiety hormones of obese rats treated with OFS, BB-12, both or neither for 8 wka
     Two way ANOVA P-values
TreatmentCOFSBB-12OFS + BB-12PreProPre × Pro
  1. a

    Values are means ± SEM, n = 9-10. Labeled means in a row without a common letter differ, P ≤ 0.05. BB-12, B. animalis subsp. lactis BB-12; C, control; CISI, composite insulin sensitivity index; GIP, glucose-dependent insulinotropic polypeptide; (Insulin AUC15)/(Glucose AUC15), ratio of insulin AUC:glucose AUC from 0 to 15 min of the oral glucose tolerance test; (Insulin AUC120)/(Glucose AUC120), ratio of insulin AUC:glucose AUC from 0 to 120 min of the oral glucose tolerance test; OFS, oligofructose; PYY, peptide tyrosine tyrosine; Pre, prebiotic; Pro, probiotic

Glucose, mmol/L X 120 min1050 ± 35.4b908 ± 24.7a917 ± 24.3a921.4 ± 21.1a0.0170.0390.012
Insulin, nmol/L X 120 min57 ± 6.645 ± 5.046 ± 2.338 ± 3.60.0350.0480.69
Amylin, nmol/L X 120 min2.30 ± 0.172.18 ± 0.092.29 ± 0.121.81 ± 0.080.0230.110.14
Leptin, nmol/L X120 min23.9 ± 2.1314.1 ± 1.5419.6 ± 2.0112.8 ± 1.050.0010.110.40
Ghrelin, nmol/L X120 min7.4 ± 1.18.5 ± 1.06.4 + 0.78.4 ± 0.80.090.560.60
PYY, nmol/L X 90 min0.75 ± 0.071.07 ± 0.100.76 ± 0.060.91 ± 0.130.0300.410.44
GIP, nmol/L X 120 min3.51 ± 0.384.04 ± 0.323.34 ± 0.293.79 ± 0.200.110.490.90
(InsulinAUC15)/(Glucose AUC15), pmol/mmol71 ± 6.064 ± 8.564 ± 7.254 ± 5.00.230.250.90
(InsulinAUC120)/(Glucose AUC120), pmol/mmol53 ± 4.950 ± 5.250 ± 3.442 ± 3.70.200.260.54
CISI, score0.51 ± 0.070.70 ± 0.050.68 ±0.040.89 ± 0.090.0040.0110.89

Insulin sensitivity and surrogate indexes of beta-cell function

Given the fact that type 2 diabetes is both a function of insulin resistance at the tissue level as well as an impairment of pancreatic beta-cell function [28], we utilized proxy measures from OGTT data to assess these parameters. Insulin sensitivity, according to CISI calculations, improved with both prebiotic (P = 0.004) and probiotic (P = 0.011) (Table 2). No effects of the treatments were seen in the early phase or total insulinogenic index calculations.

Plasma satiety hormones

For concentrations of hormones during the OGTT, there was a significant effect of time (P < 0.001) for amylin, leptin, ghrelin, GIP, and PYY; a time × prebiotic effect (P < 0.01) for GIP and PYY; and time × probiotic (P < 0.05) effect for amylin, leptin, ghrelin, GIP, and PYY (Figure 2). Amylin was reduced by prebiotic (P = 0.004) and probiotic (P = 0.063). Prebiotic reduced fasting amylin concentrations (P = 0.029; Figure 2C) and amylin AUC (P = 0.023) (Table 2). Prebiotic reduced leptin concentrations at all time points during the OGTT (P < 0.001) (Figure 2D) and leptin AUC (P < 0.001). Fasting GIP was decreased by probiotic (P = 0.02) (Figure 2F). PYY was increased by prebiotic (P = 0.001) (Figure 2G) at fasting and for AUC (P = 0.030) (Table 2).

Portal plasma gut hormones

GLP-1 was increased in portal plasma by prebiotic (P < 0.001) (Figure 2H). GLP-2 was increased by probiotic (P = 0.04) (Figure 3A). No difference in adiponectin was detected (data not shown).

Figure 3.

Portal GLP-2 (A); ileal TJP-1 and occludin mRNA levels (B); serial plasma FD-4 concentrations (C); and FD-4 AUC (D) in obese rats treated with OFS, BB-12, both or neither for 8 wk. Values are mean ± SEM, n = 9-10. BB-12, B. animalis subsp. lactis BB-12; C, control; FD-4, Fluorescein isothiocyanate-dextran - 4000 daltons; GLP-2, glucagon-like peptide-2 ; OFS, oligofructose. TJP1, tight junction protein-1.

Intestinal permeability, LPS, cytokines, and tight junction proteins

Probiotic but not prebiotic increased TJP1 mRNA levels in the ileum (P = 0.012) (Figure 3B). No differences were seen in intestinal occludin mRNA levels. In vivo intestinal permeability, tested with FD-4, did not show differences between groups (Figure 3C,D). No differences were detected in plasma LPS, TNF-α, IL-6, IL-1β, PAI-1, and MCP-1 (data not shown).

Gut Microbiota

Prebiotic increased Bacteroides spp., Lactobacillus spp., Bifidobacterium spp., and B. animalis (all: P < 0.002). Prebiotic decreased C. coccoides, C. leptum, Clostridium Cluster XI and I, and Enterobacteriaceae (all: P < 0.004) (Table 3). There was no apparent change in the composition of the gut microbiota with the probiotic. However, a direct comparison between C and BB-12 revealed that B. animalis was elevated in rats that received BB-12. This was reflected in both fecal samples (P < 0.001) and cecal matter (P < 0.001) (Figure 4). The ratio of Firmicutes (C. coccoides, C. leptum, Clostridium Cluster XI and I, Roseburia spp., Lactobacillus spp.) to Bacteroidetes (Bacteroides/Prevotella spp.) was reduced by prebiotic (P = 0.0015) and probiotic (P = 0.0011).

Figure 4.

B. animalis in fecal matter from obese rats administered BB-12 via oral gavage or in the feed. Values are mean ± SEM, n = 8-10. BB-12, Bifidobacterium animalis subsp. lactis BB-12; C, control. *Mean values of rats fed or gavaged BB-12 differ from C (P < 0.05)

Table 3. 16S rRNA copy number and relative abundance of cecal microbiota of obese rats treated with OFS, BB-12, both or neither for 8 wk a
 (×1000)Two way ANOVA P-values
TreatmentC2OFS2BB-122OFS + BB-122PreProPre × Pro
  1. a

    Values are means ± SEM, n = 9-10.

  2. b

    16S rRNA gene copies/20 ng total genomic DNA. In order to fit in the table, all values except the ratio were divided by 1000. Therefore data is 16S rRNA gene copies (103)/20 ng genomic DNA. The number in brackets indicates the relative abundance (%) of bacterial taxa per total bacteria (16S rRNA gene copies / total 16S rRNA gene copies). BB-12, B. animalis subsp. lactis BB-12; C, control; OFS, oligofructose; Pre, prebiotic; Pro, probiotic

Total bacteria43831 ± 398443103 ± 261940594 ± 248138626 ± 16980.630.170.82
Bacteroides/Prevotella spp.3747 ± 133 (8.5)5074 ± 359 (11.8)4472 ± 31 (11.0)5420 ± 462 (14.0)0.0020.130.59
Clostridium coccoides (cluster XIV)8898 ± 957 (20.3)7423 ± 389 (17.2)8115 ± 441 (20.0)6070 ± 437 (15.7)0.0040.070.63
Clostridium leptum (cluster IV)3787 ± 316 (8.6)2083 ± 117 (4.8)3516 ± 454 (8.7)1565 ± 128 (4.1)0.0010.180.67
Clostridium cluster XI0.98 ± 0.18 (0.0)0.20 ± 0.28 (0.0)0.70 ± 0.09 (0.0)0.22 ± 0.03 (0.0)0.0010.210.14
Clostridium cluster I82.3 ± 8.89 (0.2)24.0 ± 2.27 (0.1)76.3 ± 8.90 (0.2)29.1 ± 5.53 (0.1)0.000.940.42
Roseburia spp.2.48 ± 0.84 (0.0)4.78 ± 1.79 (0.0)2.86 ± 1.25 (0.0)6.00 ± 2.59 (0.0)0.140.660.81
Lactobacillus spp.2713 ± 495 (6.2)6207 ± 964 (14.4)2419 ± 272 (6.0)3899 ± 835 (10.1)0.0010.080.16
Bifidobacterium spp.234 ± 97 (0.5)3795 ± 539 (8.8)204 ± 52 (0.5)2958 ± 261 (7.7)0.0010.170.21
Bifidobacterium animalis0.030 ± 0.004 (0.0)148.5 ± 27.2 (0.3)0.82 ± 0.145 (0.0)104.4 ± 13.9 (0.3)0.000.190.19
Methanobrevibacter spp.5.44 ± 0.75 (0.0)4.14 ± 1.26 (0.0)5.08 ± 0.79 (0.0)3.96 ± 1.14 (0.0)0.250.790.93
Enterobacteriacea37.3 ± 6.4 (0.1)14.6 ± 2.4 (0.0)47.3 ± 10.8 (0.1)14.2 ± 1.3 (0.0)0.000.460.42
Firmicutes/Bacteroidetes ratio4.08 ± 0.273.18 ± 0.283.29 ± 0.312.39 ± 0.310.0040.0110.99

Two weeks of gavaging BB-12 elicited an increase in B. animalis similar to that observed with oral ingestion in the diet. B. animalis was elevated in both fecal (P < 0.001) and cecal (P < 0.01) matter in Gavage BB-12 relative to Gavage-C (Figure 4).

Discussion

Our objective was to determine if the combination of a prebiotic and probiotic enhanced the individual effects of these treatments. Based on evidence of singular antiobesogenic effects and luminal synergism between prebiotics and probiotics, we hypothesized that potentiated effects on energy intake, body composition, gut barrier function, and glycemia would be observed. OFS elicited potent effects on gut microbiota and reduced adiposity and glycemia. BB-12 improved glycemia but the effects on gut microbiota were more subtle and absent in regards to adiposity. Despite the noted metabolic improvements with OFS and BB-12, the combination of prebiotics and probiotics did not yield potentiated host metabolic responses.

Classically, prebiotic feeding has been associated with a reduction in energy intake and body fat [18]. Our study confirms that OFS supplementation elicits a reduction in energy intake, body fat, and weight gain. OFS contributes to reduced energy intake by decreasing the overall caloric density of the diet and by enhancing satiety through the secretion of satiety hormones GLP-1 and PYY [7, 18, 29, 30]. In support of this purported mechanism, we observed that OFS increased fasting portal plasma GLP-1 as well as PYY AUC.

Evidence for the ability of probiotics to reduce body weight and adiposity is more limited and controversial [10]. Our study indicates that BB-12 does not affect body weight or adiposity. A recent study in DIO Sprague Dawley rats examined the effect of four different strains of Bifidobacterium spp. on body weight. Of the four strains tested, one was found to reduce weight gain, whereas the other three were weight neutral or increased weight gain [15]. Two studies recently reported that B. breve and B. adolescentis limit weight gain and adipose tissue mass in DIO animal models [12, 13]. Other studies have found that Lactobacillus gasseri, in both animals and humans, reduces body weight and adiposity [11, 31]. Altogether, this evidence suggests that probiotic effects on body weight may be strain and model dependent.

Bifidobacterium longum ssp. infantis and other bifidobacteria are some of the first organisms to colonize the gut after childbirth and are recognized as an important commensal group of bacteria [32]. Our study demonstrated the classical “bifidogenic” effect of OFS [33], with Bifidobacterium spp. and B. animalis increasing relative to C. Interestingly, the increase in B. animalis elicited by OFS far surpassed that of BB-12. Contrary to our hypothesis, probiotic did not produce an independent effect for increasing Bifidobacterium spp. or B. animalis when analyzed against prebiotic. Relative to C, BB-12 did elicit an increase in B. animalis but this increase was low compared to the increase observed with the prebiotic. Given these results, we were concerned that BB-12 had been rendered nonviable in the diet matrix. We therefore completed an additional study to determine if administering the BB-12 via oral gavage would increase intestinal B. animalis. Virtually no differences in fecal B. animalis concentrations were observed with gavage versus feed-administered BB-12. Our findings are consistent with that of another study which found that provision of B. animalis ssp. lactis GCL2505 increased B. animalis ssp. lactis but no other endogenous bifidobacteria [34]. These findings likely reflect the vulnerability of probiotics to the harsh environment of the digestive tract [34].

In some but not all reports, obesity has been associated with a decrease in Bacteroidetes and a proportional increase in Firmicutes [5, 35]. Despite no main effect of probiotic on any one genus or species of bacteria, both prebiotic and probiotic decreased the ratio of Firmicutes to Bacteroidetes. This is relevant in that an obesity associated gut microbiota has been found to be more efficient at extracting luminal energy, potentially contributing to weight gain [36]. Prebiotic decreased the ratio between Firmicutes and Bacteroidetes because of an increase in Bacteroides spp., and a decrease in prominent members of Clostridium spp., including C. leptum and C. coccoides. Lactobacilli, a group of bacteria belonging to the Firmicutes phylum, also increased in the prebiotic groups, which is important to note as certain lactobacilli have been recognized for health promoting properties. In our study probiotic BB-12 did not increase Bacteroides spp. or Lactobacillus spp. With Lactobacillus spp. belonging to the Firmicutes phylum, the lack of change on Lactobacillus spp. with probiotic contributed in part to a reduced ratio of Firmicutes to Bacteroidetes. Therefore, despite seeing beneficial improvements in the ratio between Firmicutes to Bacteroidetes with probiotic treatment, this does not necessarily reflect a more advantageous profile of bacteria with the probiotic relative to the prebiotic.

In type 2 diabetes, hyperglycemia is considered the primary factor propagating metabolic decline [28]. For this reason, strategies that improve glucose control are imperative for successful management of the disease. Our study identified an interaction effect between prebiotic and probiotic on glucose levels during an OGTT. The interaction between prebiotic and probiotic, however, was not a potentiated reduction in glycemia. Rather, OFS, BB-12, and OFS+BB-12 all reduced glycemia to the same extent, suggesting that the dose of prebiotic or probiotic alone saturated the effect. Independently, both prebiotic and probiotic reduced postprandial insulin with the probiotic also reducing insulin levels at fasting. The CISI calculation, with larger numbers indicating improved insulin sensitivity, indicates that both prebiotic and probiotic improved insulin sensitivity. This is an important consideration given the fact that insulin resistance can place additional stress on beta-cells in the pancreas in the context of hyperglycemia [37]. While an improvement in glucose control has been reported in overweight and obese adults consuming prebiotics [38], the current data highlights a potential role for probiotics in improving glycemia that should be verified in humans as well.

Gut barrier function plays an important role in halting the development of diabetes and metabolic disease, in part by limiting the passage of pro-inflammatory agents from the gut into the circulation. LPS, a lipoglycan found on the cell surface of gram negative bacteria, interacts with toll-like receptor 4 (TLR4) and initiates the release of proinflammatory cytokines [39, 40]. GLP-2, a peptide that is co-secreted with GLP-1, plays a pivotal role in intestinal adaptation, epithelial cell-proliferation, and maintenance of gut integrity [20]. Previous studies have found that GLP-2 elicits improvements in gut permeability, endotoxemia, and increased expression of zonula occludens (ZO-1) and occludin [20]. We observed an increase in GLP-2 specifically with probiotic. While we did not see changes in intestinal permeability with the FD-4 experiments, LPS, or pro-inflammatory cytokines in plasma, we did find that probiotics elevated mRNA TJP1 expression.

Our study was limited by the fact that the rats gained less fat mass than has typically been seen in our DIO model. Following similar durations of high energy diet consumption, our DIO rats typically display body fat percentages ranging from 17% to 24% which is lower than the 10-13% seen in the current study. This lower body fat % is one possible explanation why we did not find the significant changes in gut barrier integrity and endotoxemia that has been observed in other studies with ob/ob mice [20, 21]. Furthermore, our microbial 16S rRNA analysis does not allow us to understand the function of the identified bacteria and future studies using a microbial metagenomic sequencing approach would provide valuable information about the functional diversity of the bacterial community.

In conclusion, despite the hypothesized synergistic effects between prebiotic and probiotic, little potentiation of the effects was observed here in the case of OFS and BB-12. Our results demonstrate that prebiotics, in comparison to probiotics, provide a more potent stimulus in reducing adiposity and modifying gut microbiota. Both prebiotics and probiotics, however, offer benefits in regards to improving glucose tolerance. With a need for more noninvasive strategies for individuals with overweight and obesity, both prebiotic and probiotic treatments offer promise for improving metabolic outcomes, particularly glycemia.

Acknowledgments

The authors would like to thank Chr. Hansen (Milwaukee, WI) for the generous donation of Bifidobacterium animalis subsp. lactis BB-12 for our study. Kristine Lee, Faculty of Kinesiology, University of Calgary for technical assistance. Carol Stremich, Biological Sciences, University of Calgary for technical assistance. Dr. D. Morck, Biological Sciences, for allowing us to use his anaerobic hood. Dr. L. McMullen and Dr. M. Gaenzle, Agricultural, Food and Nutritional Science, University of Alberta for technical assistance.

Ancillary